Cancer Stem Cells (CSC) represent a very small population within in vitro cell lines and in situ tumors. Currently, isolation can only be appropriately done with FLOW Cytometry cell sorting capabilities. Additionally, tissue culture methods have been devised to maintain stem cell characteristics to prevent differentiation into progeny cells. We have been able to charaterize CSC from various epithelial carcinoma cell lines. 1) We have identified the CSC phenotype of prostate cancer cell lines, which agrees with clinical tissue observation, as a CD44+, 24- subpopulation, these cells are genomically distinct from other populations and uniquely develop tumors in murine xenograft models. 2) We have developed in vitro methods to study how CSC differentiate into progeny cells, a phenomenon which occurs in situ. We have tentatively identified the extrinsic substance which induces differentiation and are developing methods to block this pathway in vitro and in CSC xenografts. 3) Identification of immunosuppressive/toleragenic markers on cancer stem cells. We have found that almost all CSC found in melanoma, prostate, breast, and renal cell carcinomas have an immunosuppressive surface antigen called CD200, suggesting that the immune system may recognize or not CSC differently from the progeny cells. 4) The effects phytochemicals on CSC viability. Because of the historical research regarding the effects of certain phytonutrients on cancer prevent, we screened a highly selected libraray for induction of apoptosis in CSC compared to normal stem cells. WE have found two that kill CSC, their progeny, but have little effects on normal stem cells. We have discovered three phytochemicals that block CSC growth in xenografts and determining the mechanism(s) by which the phytochemicals work. 5) We have identified novel characteristics in polycomb gene expression which controls docetaxel sensitivity in prostate cancer.Also findings that highly invasive pacreatic cancer cells exhibit increased expression of DNA repair genes.